Charles Hoskinson Says New Cardano (ADA) Language Will Be First to Pair Intents With Groth16 Proofs

Intents and Groth16 Proofs in One Language

A developer writing for Cardano (ADA) will, if the network's founder has his way, be able to state what a transaction is meant to do and attach cryptographic proof of it in the same place. Charles Hoskinson said on October 18 that he is building a new programming language for Cardano, one designed to process intent structures together with proof carrying data, with verification running through Groth16 proofs. Intent structures let a transaction declare its intended outcome instead of its low-level mechanics. Proof carrying data is a format in which a piece of data arrives with its own verifiable evidence attached. Groth16 is one of the most widely used proving systems in the zero-knowledge field. Hoskinson claims the structure will be the first to combine all three, a combination he argues could open several new use cases across the Cardano ecosystem. In practical terms, a contract written this way could let an application deployed on the mainnet demonstrate to anyone checking that a transfer or computation followed the rules its author declared. Hoskinson has presented advanced cryptography as a headline feature of the language, without elaborating on which parts of the stack it will touch. The statement came as a reply to a community member who asked what unique advantages Cardano (ADA) offers as a highly decentralized base layer. Hoskinson answered that the work under way could strengthen the network's technical case, adding weight to the argument that a proof-of-stake chain built to this level of decentralization carries benefits no other design replicates. He gave no name for the language, no technical specification and no timeline. The answer matters for Cardano's positioning, since the question it responded to went to the core of what separates the chain from competing base layers. The Cardano (ADA) price has held near $0.22 through an 11% weekly decline, and an announcement without code attached is unlikely to move that picture on its own.

Haskell Doubts and the Midnight Concept

The announcement reopens a question Hoskinson has returned to repeatedly: which language should Cardano (ADA) actually be built on? The network's development architecture has rested largely on Haskell since its early stages, and Hoskinson defended that choice in public for years. Haskell is a functional programming language known for strong static typing, a fit Hoskinson long argued suits financial contracts where correctness matters more than convenience. In June 2022 he joined the exchange between Elon Musk and Jack Dorsey over the merits of programming languages, and after Musk spoke of his interest in C, Hoskinson suggested Haskell deserved a try. He later conceded, however, that he was never fully certain about the decision to build Cardano with Haskell. By May 2024 he said he had seriously weighed Clojure as an alternative, with the availability of Cloud Haskell and the ability to draw on tools such as Liquid Haskell and Agda ultimately proving decisive. That deliberation is the clearest earlier sign his attachment to Haskell was conditional. In an interview in April 2025 he went further, saying that if Cardano could be rewritten from scratch he would probably pick a different language. Read against that record, a language developed by Hoskinson personally reads as the first concrete step beyond deliberation. That matters because a language handling intents natively could compress what today takes multiple layers of tooling into a single development path. Separately, he said he is working on a concept he believes can solve most of the problems in crypto, one he argues only the Midnight Network can bring to the space, and one he considers possibly the most important work of his career. Midnight is the privacy-focused project inside the Cardano ecosystem. Whether that concept and the new programming language are connected was left open in his remarks. He offered no indication of when either would surface in usable form.

No Name, No Release Date

Our reading is that this is a direction claim, not a shipping event. Hoskinson has tied the network's technical completion to the Leios roadmap, with the testnet running at 26.8 txkB/s, so a new language would sit alongside that work rather than replace it. Nothing in the announcement is verifiable yet: no name, no specification, no code and no stated link to the Midnight concept. Speculation around the asset continues elsewhere, from the Cardano technical analysis side to a Bodega Market Cardano stablecoin wager on a $1 million target. Until something ships, developers still build on the Haskell toolchain, which is the constraint that remains.